Re: Painter Fuzzy Node in github

[Gregory Casamento]

Johan,

The URL http://github.com/gnustep is my account for the mirror from
SVN.   I think your URL is incorrect.

GC

On Wed, Dec 17, 2014 at 2:00 PM, Johan Ceuppens <address@hidden> wrote:
> Hi,
>
> I took out the fuzzy out of gnustep-fuzzy and am trying to make it work by
> reducing warnings. There's only prolog-decision tree code in the makefile
> now as this is where Ivan sent me to CALayer updating and wont a tree and
> said the thing should not be too cumbersome.
>
> 2014-12-17 17:34 GMT+01:00 David Chisnall <address@hidden>:
>>
>> On 17 Dec 2014, at 17:23, Johan Ceuppens <address@hidden> wrote:
>>
>> > Hello,
>> >
>> > I am writing a small prolog system with a decision tree which
>> > understands "not x" and "x" for now. This can be used in the CALayer as 
>> > said
>> > before by Ivan here.
>> >
>> > The interface is as follows :
>> >         //make predicates which is an NSString
>> >         OpalFuzzyPredicate *pred = [[OpalFuzzyPredicate alloc] init];
>> >         //initialize it with a string which in this case is multi-worded
>> >         [pred initWithString:@"update full window";
>>
>> Sending an init-family method to an object that has already been
>> initialised is usually undefined behaviour.
>>
>
> I took everything out that way.
>
>>
>> >        //so it becomes a compound, which gets parsed and
>> >         //adds comp to compound DB and compiles it into the tree
>> >         InferenceCompound *comp = [factory makeCompound:pred];
>>
>> What is factory?  In Objective-C classes are usually factories, there's
>> little need for the factory pattern in a language where class methods can be
>> overridden.
>>
>>
>> >
>> >         pred = [[OpalFuzzyPredicate alloc] init];
>> >         //initialize it with a string which in this case is multi-worded
>> >         [pred initWithString:@"not update full window";
>>
>> Once again.  And why would you want to do this rather than reuse
>> NSPredicate, which can already represent and operate on predicates?
>>
>
> Still have to look that up but I am not sure I want to use it as my
> predicate needs to be small at the handle.
>
>>
>> >
>> >         //so it becomes a compound, which gets parsed and
>> >         //adds comp to compound DB and compiles it into the tree
>> >         InferenceCompound *comp = [factory makeCompound:pred];
>> >
>> > I will try to get this in if I can do it.
>>
>> Given your unwillingness to fully explain what you're doing and the
>> appalling quality of your code (some of the diffs you've posted obviously
>> wouldn't even compile due to mismatched braces, the rest is broken due to
>> poor understanding of Objective-C conventions), I'd be very hesitant about
>> your having commit access to a GNUstep repository until you've demonstrated
>> that:
>>
>
> I implement by heart, I do not enter repetitive debugging phases.
> diffs are out, I upload to github, see below.
>
>>
>> - You can communicate with other developers
>
>
> Which I am doing.
>
>>
>> - You can write useable code
>
>
> Tryout http://github.com/gnustep/gnustep-fuzzy it should work.
>
> Best regards,
> `Enry
>
> _______________________________________________
> Gnustep-dev mailing list
> address@hidden
> https://lists.gnu.org/mailman/listinfo/gnustep-dev
>



-- 
Gregory Casamento
GNUstep Lead Developer / OLC, Principal Consultant
http://www.gnustep.org - http://heronsperch.blogspot.com
http://ind.ie/phoenix/